Clase1b-Introduccion_Python_preclase.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Clase 1b: Estructura de datos y funciones en Python\n",
    "Referencia: Capítulo 3 de \"Python for Data Analysis\", Wes McKinney."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Estructuras de Datos\n",
    "Aparte de una variable que tiene un sólo valor, Python posee múltiples estructuras de datos para almacenar información. Estas son algunas de las más básicas que podríamos necesitar."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tuplas (`tuple`)\n",
    "Una tupla es un conjunto fijo de elementos con un orden dado. Se definen con parentesis redondos `(`, `)` o simplemente sin ellos, separando los elementos por coma."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tup = (4, 5, 6)\n",
    "tup"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Al igual que los tipos básicos,  podemos convertir otros objetos en tuplas con el comando `tuples`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tuple([4, 0, 2])\n",
    "tup = tuple('perro')\n",
    "tup"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para acceder a u elemento de una tupla, usamos parentesis cuadrados.\n",
    "\n",
    "**RECORDAR: en python en general las cosas se numeran desde 0**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tup[1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "También las operaciones básicas aplican sobre las tuplas. Por ejemplo, la suma `+` se considera como la concatenación."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "(4, None, 'foo') + (6, 0) + ('bar',)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "('foo', 'bar') * 4"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "y, como todo objeto en Python, tienen sus propios metodos (en este caso, `.count` y `.index`)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "a = (1, 2, 2, 2, 3, 4, 2)\n",
    "a.count(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Listas (`list`)\n",
    "Similar a la tupla, pero está pensado para que su tamaño sea variable. Es decir, agregar y quitar elementos a la lista. Se define como una tupla pero con paréntesis cuadrados `[` `]`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "[2, 3, 7, None]\n",
    "tup = ('foo', 'bar', 'baz')\n",
    "b_list = list(tup)\n",
    "b_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "b_list[0] =65\n",
    "b_list."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "gen = range(10)\n",
    "gen\n",
    "list(gen)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "b_list.insert?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Se pueden agregar elementos (`.append`), insertar elementos en una posición específica (`.insert`) o sacar elementos con un índice en particular (`.pop`) o sacar un elemento por su valor (`.remove`)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "b_list.append('dwarf')\n",
    "b_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "b_list.insert(1, 'red')\n",
    "b_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "b_list.pop(2)\n",
    "b_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "b_list.append('baz')\n",
    "b_list.append('chao')\n",
    "b_list\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#b_list.append('foo')\n",
    "#b_list\n",
    "b_list.remove('baz')\n",
    "b_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "b_list.remove?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "También es facil ver si un elemento está en una lista (comando `in`). Por ejemplo, para un ciclo `for`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "'dwarf' in b_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "'dwarf' not in b_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "[4, None, 'foo'] + [7, 8, (2, 3)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Ordenar (`.sort`)\n",
    "La listas se pueden ordenar"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "a = [7, 2, 5, 1, 3]\n",
    "a.sort()\n",
    "a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "b = ['saw', 'small', 'He', 'foxes', 'six']\n",
    "b.sort()\n",
    "b"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Tambien podemos pasarle una función para determinar como ordenar. Por ejemplo, el largo del string (`len`)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "len('hola amigo')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "b.sort(key=len)\n",
    "b"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Slicing\n",
    "Similar a otros lenguajes (como Matlab), Python permite selecciona un _slice_ de la lista usando la notación `inicio:final`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "seq = [7, 2, 3, 7, 5, 6, 0, 1]\n",
    "seq[3:4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "seq[3:4] = [6, 3]\n",
    "seq"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Si omito el inicio o final, se asume"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "seq[:5]\n",
    "seq[3:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Si uso un número negativo, se hace relativo al otro extremo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "seq[-4:]\n",
    "seq[-6:-2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "O por ejemplo, de dos en dos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "seq"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "seq[::-2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "(aunque posiblemente es mas natural usar la función `reversed`)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "list(reversed(seq))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Diccionarios (`dict`)\n",
    "Esta es una gran novedad de python. Permite asociar un llave (_key_) a un valor (_value_). Permite hacer una especie de lista, pero asociado a una llave en particular.  Se definen con llaves `{` `}`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "empty_dict = {}\n",
    "d1 = {'nombre' : 'Eduardo', 'edad' : 43}\n",
    "d1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "d1['edad']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para acceder a un elemento, igual que las tuplas y listas, se usa parentesis cuadrado. Pero ojo que a diferencia de antes, no hay un \"orden\", sino uno accede a la llave (`key`) indicada"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "d1['nombre']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "d1['siete'] = 'un entero'\n",
    "d1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "d1['apellidos'] = ['Moreno','Araya']\n",
    "d1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "d1['apellidos']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "d1['apellidos'][1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "d1['apellidos'][1].upper()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "d1.values()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Como dijimos, un diccionario tiene llaves y valores. Se puede acceder a ellos con `.keys()` y `.values()`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "list(d1.keys())\n",
    "list(d1.values())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Puedo actualizar el diccionario ya sea cambiando un elemento especifico o con otro diccionario"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "d1['edad']=43\n",
    "d1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "d1.update({'b' : 'foo', 'nombre' : 'Carlos'})\n",
    "d1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nombres = ['Sebastian', 'Juan', 'Pedro']\n",
    "apellido = ['Perez', 'Martinez', 'Perez']\n",
    "dic = {apellido[i] : nombres[i] for i in range(3)}\n",
    "dic"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conjuntos (`set`)\n",
    "Como sabemos de álgebra, un conjunto es una coleccion de elementos _únicos_, no ordenados. Se puede definir con llaves `{` `}` o con el comando `set`. Recordar! los elementos son únicos en un conjunto."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "set([2, 2, 2, 1, 3, 3])\n",
    "{2, 2, 2, 1, 3, 3}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "a = {3,4,6,2,1,4,3,2,3,2}\n",
    "a"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Todas las operaciones básicas de un conjunto están definidas: `.add`, `.remove`, `.union` (`|`), `.intersection` (`&`), `.difference` (`-`), y otras."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "a = {1, 2, 3, 4, 5}\n",
    "b = {3, 4, 5, 6, 7, 8}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "a.union(b)\n",
    "a | b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "a.intersection(b)\n",
    "a & b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "{1, 2, 3} == {3, 2, 2, 3, 1}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Estructuras en notación abreviada o expresiones ternarias\n",
    "Como dijimos, otra ventaja de Python es poder usar `for` of `if` para definir en forma abreviada una lista, tupla o diccionario.\n",
    "\n",
    "Por ejemplo, para una lista,  notación es:\n",
    "```[<expresion> for <valor> in <coleccion> if <condicion>]``` "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "strings = ['a', 'as', 'bat', 'car', 'dove', 'python']\n",
    "strings\n",
    "strings[0].upper()\n",
    "[x.upper()  for x in strings]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "[x.upper() if len(x)>2 else x for x in strings]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para diccionarios: `{<llave-expr> : <valor-expr> for <valor> in <coleccion> if <condicion>}`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "{val : len(val) for val in strings if val != 'car'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "len('perro')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Funciones"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Como todo lenguaje, podemos definir funciones propias, que reciben objetos como entrada, y retornan un objeto como salida.  Estas se definen con el comando `def`\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def suma(x,y):\n",
    "    return x+y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "suma(4,6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def suma(x,y=1):\n",
    "    return x+y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "suma(4)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "El orden de los parametros es importante. También podemos agregar parametros con un valor por defecto."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def my_function(x, y, z=1.5):\n",
    "    if z > 1:\n",
    "        return z * (x + y)\n",
    "    else:\n",
    "        return z / (x + y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "my_function(5, 6, z=0.7)\n",
    "my_function(3.14, 7, 3.5)\n",
    "my_function(10, 20)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Los parametros de las funciones son locales, o sea, se modifican dentro de la función."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def func(a):\n",
    "    for i in range(5):\n",
    "        a.append(i)\n",
    "    return a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "b=[2,3]\n",
    "func(b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "a"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "También una función puede devolver multiples valores o salidas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def f(x):\n",
    "    out1 = x+1\n",
    "    out2 = x/2\n",
    "    out3 = x**2\n",
    "    return out1, out2, out3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "f(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "a, b, c = f(2)\n",
    "c"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**EJERCICIO**: Genere una función que dado un número, retorna la suma de los números del 1 a ese número, que no son divisibles por 7."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def mi_ejercicio(hasta):\n",
    "    suma = 0\n",
    "    for i in range(hasta+1):  # Recordar que range no incluye el ultimo número.\n",
    "        if i%7 != 0:\n",
    "            suma = suma + i\n",
    "    return suma"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "mi_ejercicio(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Funciones anónimas  (Lambda Functions)\n",
    "Otra gracia de Python es poder expresar en forma compacta una función. Por ejemplo, en vez de definir:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def short_function(x):\n",
    "    return x * 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "podemos usar la notacion compacta:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "equiv_anon = lambda x : x*2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "short_function(3)\n",
    "equiv_anon(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "El nombre \"funcion anónima\" viene del hecho que no necesitamos darle un nombre como en `def`, sino que la usamos para definir algo en solo una linea. Esto es muy util trabajando con datos, porque es muy comun querer aplicar una función a todos los elementos de un arreglo."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "por ejemplo, antes usamos ordenamos un conjunto de elementos por su largo pasando como argumento la función `len()`. Ahora podemos hacer cosas mas complejas, por ejemplo, el numero de elementos distintos."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "strings = ['foo', 'card', 'bar', 'aaaa', 'abab']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "strings.sort(key=len)\n",
    "strings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def mifunc (s):\n",
    "    return len(set(list(s)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "mifunc('aaaa')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "strings.sort(key=mifunc)\n",
    "strings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "strings.sort(key=lambda x: len(set(list(x))))\n",
    "strings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.7.7 64-bit",
   "language": "python",
   "name": "python37764bitbf82ff8733e44256b381353da7db86a3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}